Content uploaded by Sinem Güdüm
Author content
All content in this area was uploaded by Sinem Güdüm on Nov 18, 2020
Content may be subject to copyright.
DİYALEKTOLOG
ULUSAL SOSYAL BİLİMLER DERGİSİ YAZ SUMMER 2020 -SAYI /ISSUE 24- SAYFA PAGE 93-101
93
USING ANTHROPOMORPHISM IN MOVIES FOR PROMOTING HUMANOID
ROBOTS IN THE AGE OF ARTIFICIAL INTELLIGENCE: EX-MACHINA MOVIE
Sinem GÜDÜM
1
Abstract
In the Merriam-Webster dictionary (2020), humanization is described as representing something
as “human” by giving it human qualities. This makes the non-human seem gentler, kinder, or
more appealing to people. Such a trial to attribute some human characteristics to non-humans is
defined as “Anthropomorphism”. The results of studies show that a product presentation with
anthropomorphism can actually be an influencial advertising tool, since it creates positive
emotions among the audience and has a positive affect in their purchasing behavior. But are there
any other possible underlying reasons for such an increase in sales when consumers are made to
perceive material objects as ‘human’? And is this advertising strategy being used in the movies
today to promote the use of humanoid-robots? It should be noted that this kind of correlation made
between the object and the subject (non living and the living) may have foreseeable drawbacks in
the future, and they need to be attended. In this article, using anthropomorphism in movies for
promoting humanoid robots will be discussed by analyzing the “EX-MACHINA” movie as a
case study, and applying content and discourse analysis on the subject.
Keywords: Anthropomorphism, Interactivity 2.0, Avatar Brands, Artificial Intelligence,
Advertising
YAPAY ZEKA ÇAĞINDA İNSANSI ROBOTLARIN TANITIMI İÇİN FİLMLERDE
ANTROPOMORFİZM KULLANIMI: EX-MACHINA FİLMİ
Özet
Merriam-Webster sözlüğünde (2020), “insanlaştırma / humanization” bir şeyin “insan” olarak
temsil edilir hale sokulması olarak tanımlanır; başka bir deyişle, insani nitelikleri insan olmayana
atfedilmesidir. Cansız objelere, hayvan ya da bitkilere insani özellikler vermek, onların bireyler
tarafından daha ‘bilindik’ algılanmalarına ve dolayısıyla daha yakın gözükmelerine sebebiyet
verebilir. Tek başına bu tanım, insan yaşamında ve dolayısıyla reklamcılıkta "nesne" nin sürekli
insanileştirilmesinin arkasındaki nedenlerden birini göstermektedir. İnsan özelliklerini insan
olmayanlara atfetmek, antropomorfizm (insanbiçimcilik) olarak tanımlanır. Çalışmaların
sonuçları gösteriyor ki, insanlaştırılmış bir ürün sunumunun izleyiciler arasında olumlu duygular
yaratması ve satın alma davranışlarında olumlu bir etki yaratması nedeniyle antropomorphizm
etkili reklam stratejilerindendir. Bu bağlamda, tüketicilerin nesneleri 'insan' olarak algılamaları
sağlandığında satışlarda artış görülmesinin altında yatan başka olası nedenler nelerdir? Yapay
zeka çağında insansı robotların kullanımını teşvik etmek için filmler bir reklam aracı olarak mı
kullanılmaktadır? Nesne ile özne arasında kurulan böylesi korelasyonların gelecekte öngörülebilir
sakıncaları olabileceği açıktır. Bu makalede, insansı robotları tanıtmak için filmlerde
1
Dr. Öğr. Üyesi, Marmara Üniversitesi, İletişim Fakültesi, Reklamcılık ve Tanıtım-İstanbul
e-posta: sinem.gudum@marmara.edu.tr
DİYALEKTOLOG
ULUSAL SOSYAL BİLİMLER DERGİSİ SİNEM GÜDÜM
YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
94
antropomorfizmin kullanılması, “EX-MACHINA” filmi bir vaka çalışması olarak ele alınarak
yapılacak; konu, ilgili içerik analizi üzerine söylem analizi uygulanarak tartışılacaktır.
Anahtar kelimeler: Antropomorfizm, etkileşim 2.0, Avatar Markaları, Yapay Zeka, Reklam
INTRODUCTION
Attributing human characteristics to non-humans (products, objects etc…) is known as
anthropomorphism (Guthrie, 1993:62). The results of many studies show that a product
presentation with anthropomorphism can actually be a very influencial advertising strategy since
it elicitates positive emotions among the audience. It has an emotional appeal which creates a
positive affect in the purchasing behavior (Klein, 2013:265). Object (objectum, gegenstand,
objet..), in ethimologic terms, stands for the things that exist on the “outside”, things which stand
against the “subject” (Bilgin,N.2011:8-9). So, humanization is indeed an effort in familiarizing
(and thus bringing closer) the subject with the object by re-defining the distinct division between
the two.
Anthropomorphism is especially used in advertising throughout the world. From Coca-
Cola with its original bottle shaped like a curvy woman, to chatbots wih human names (Siri,
Alexa…), and from world famous anthropomorphic Lightning McQueen in the ‘Cars’ movie, to
the human-like C-3PO in Starwars, both of which were used in the advertising campigns. Reavey
et al. (2018:440) concludes that consumers mostly prefer watching advertisements that use overt
humanization (instead of subtle, or no-humanization) when an ad promotes assertive language
(i.e. Just Buy!); however, the same consumers, prefer ads that use more subtle humanization when
there is no assertive language. Can this be regarded as an unconscious reaction by the humans to
stay away from taking direct orders from the ‘non-human’ and actually to keep their power over
any given situation by reminding themselves who is actually in “charge”?
Ex-Machina movie starts with a very simple but profound statement which is worth
mentioning: “There’s nothing more human then the will to survive.” When the human feels that
his/her life is threatened, then the so called “animal instinct” comes on the stage. Horkheimer,
and Adorno (2019:15-16), touched upon this concept in their work “Towards a New Manifesto”
and said: “Happiness would be an animal condition viewed from the perspective of whatever has
ceased to be animal. And that to achieve the condition of an animal at the level of reflection would
be freedom.” So, can one say that ‘the animal’ side of the human feels safer by using
anthropomorphism and familiarizing the reflection?
Riek and friends (2009:245) stated that people empathize more strongly with humanoid
robots and less with mechanical- looking robots. This is explained by Masahiro Mori’s (2012:98)
theory called the “Uncanny Valley Effect”, which states that as robots become more ‘humanlike’,
it makes them more ‘familiar’ for people, and thus more ‘likeable’. Mori calls this feeling ‘a
heightened sense of affinity’. But there comes a time when the mismatch between the two forms
(humans vs humanoid robots), the emphaty lacking ‘interactivity’, and motion quality start
elicitating a sense of unease in the human. Mori gives the example of shaking a prosthetic hand
with a high resemblance to the human form. He says, once a person realizes that the hand is
Using Anthropomorphism in Movies For DİYALEKTOLOG
Promoting Humanoid Robots in The Age of ULUSAL SOSYAL BİLİMLER DERGİSİ
Artificial Intelligence: YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
Ex-Machina Movie
95
artificial (because of its coldness, or its texture etc …), he /she experiences an eerie sensation, and
looses his/her sense of affinity; this is the moment when the hand becomes ‘uncanny’.
This notion can also be analyzed by the ‘Simulation Theory’, which suggests that humans
understand the minds of others through creating “simulations” for emphatisizing with another’s
situation to understand them (Riek et al.2009:245). Here it may be appropriate to discuss another
academic writer’s, namely Roy Ascott’s (2016), words: “The real was only virtual after all. Now
we have a different paradox to deal with-actually to celebrate: the virtual is becoming real”.
If this statement is true, then one may really become the “Human component in a Turing
Test” as mentioned in the movie Ex-Machina. Some academic research relevant to the object -
subject affinity can be summarized as below:
RESEARCHERS
YEAR
FINDINGS
Caporael
1986
Identifying with an object, induces high affect,
accompanied by anthropomorphic behavior.
Goetz,Kiesler&Powers
2003
People prefer more humanlike robots for jobs re-
quiring more sociability.
Hinds,Roberts & Jones
2004
People have the tendency to take less credit for the
work done and also less personal responsibility as
robots become more humanlike
Krach,Hegel,Wrede,Sagerer,
Binkofski & Kircher.
2008
a strong linear relationship can be spotted between
cortical activation regarding how humans process
other minds, and anthropomorphization.
Chandler and Schwarz
2010
Consumers who likes to anthropomorphize their
vehicles, seem to be less willing to replace them.
1. INTERACTIVITY 2.0 AND CONSUMER: CHAT WITH CYBORGS AND BOTS
Roy Ascott (2016) says that reality has always been constructed and mediated by the
ultimate technology, human language, in all its varied configurations. And interactive
telecommunications is also a language before it is anything else. Thus, interactive
telecommunications is a technology that empowers the individual to connect with others.This,
however, also has a downside: Today, everything in our online world is indexed. Every search
that one makes, feeds the Google database; every ‘like’ that one gives, helps Facebook to give
better targeted advertising solutions to its customers; every story someone posts, makes Instagram
a better choice for both the entertainment seekers, and the influencers... So what if Google, and
the like, or any of the social media companies that we entrust our information with, wanted to use
our search histories in order to create the most artificially intelligent robot? Would the humans
still hold the comparative advantage thank to their consciousness? Can a mutually shared meaning
DİYALEKTOLOG
ULUSAL SOSYAL BİLİMLER DERGİSİ SİNEM GÜDÜM
YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
96
amongst people, or amongst any other communicating/interacting intelligence holders, be
possible without consciousness?
Interactivity has been a term used for describing a natural attribute of face-to face
conversation, together with some mediated communication settings like enabling interactivity
between two-way cable systems, electronic text systems, and some programming work.
Interactivity was, and still is, also present in the operation of traditional media. Examples of this
can be given as: Letters to the editor, talk shows and audience participation on television and
radio programs. However, for full interactivity to occure, communication roles need to be
interchangeable: role assignment and tum-taking are to be non-automatic or nearly so
(Rafaeli,2018:109). But one may say that starting from the times of web 2.0, and especially as we
approach the Age of Artificial Intelligence, the definition of ‘interactivity’ has become
questionable, and insufficient. Can one call a chat with a robot, or a chatbot interactivity? If we
name this new era of communication as ‘Interactivity 2.0’, then ‘who’ or better to say ‘what’ will
one be actually communicating with?
In the 2015 movie ‘EX-MACHINA’, a brilliant scientist, Nathan, brings a successfull
programmer, Caleb, to his well-hidden cabin which operates as a research facility. Nathan’s aim
(or at least the aim that he told Caleb) is to test if his new invention, a cyborg called Ava, truly
has consciousness just like a humanbeing. At this point, a scene from ‘Ex Machina’ where Nathan
Bateman talks about Artificial Intelligence (AI) comes to mind:
Nathan: “One day the AIs are going to look back on us the same way we look at fossil
skeletons on the plains of Africa. An upright ape living in dust with crude language and tools, all
set for extinction.”
Here, how one defines the ‘human’ becomes utterly important; Is human merely a “body”,
or is the human’s most deterministic characteristic his/her consciousness? If the human
consciousness shapes the AI, can it be actually considered as partly ‘human’? And at this point
one important question comes to mind: Can any living being without a ‘free-will’ be considered
as human?
Human-computer interaction has its roots in the Turing test, where a human interacts with
a computer. If the human doesn’t understand that he/she is interacting with a computer, the test is
passed. In the movie, the Turing test is explained as below (min:10:26):
Nathan: Do you know what the Turing Test is?
Caleb: Yeah. I know what the Turing Test is. It’s where a human interacts with a computer.
and if the human can’t tell they’re interacting with a computer, the test is passed.
Nathan: And what does a pass tell us?
Caleb: That the computer has artificial intelligence.
Using Anthropomorphism in Movies For DİYALEKTOLOG
Promoting Humanoid Robots in The Age of ULUSAL SOSYAL BİLİMLER DERGİSİ
Artificial Intelligence: YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
Ex-Machina Movie
97
The chatbots that we come across on e-commerce websites and/or shopping apps today,
can be given as an example for such a human-AI interaction. At this point it would be relevant to
talk about some of the marketing campaigns that were made especially for AVA at the time of
the Ex- Machina movie (2015):
• AVA had a Linkedin account as a humanoid robot, promoting herself as a personal
assistant. The account is closed now.
• The marketing team from the film created a Tinder profile for the humanoid lead-
character Ava, who was played by Alicia Vikander. The design and positioning of the
campaign matched perfectly with the story of the film. Marketing crew used photos of
Vikander, and created a Tinder bot that asked her visitors some questions on what makes
a person human, would they like to meet and what attracts them to her... After answering
those trivia questions, the user was directed to an Instagram account that promoted the
film. Many people were charmed by the sweet face of the bot AVA online, and they
wanted to ‘communicate’. Here, one may quote Caleb’s words about AVA in the movie:
“To escape, she would have to use imagination, sexuality, selfawareness, empathy,
manipulation, and she did. If that isn’t AI…” Actually Caleb was right, people couldn’t
escape from AVA’s ‘bot’ charms.
(https://mashable.com/2015/03/16/ex-machina-tinder-marketing/ 07.05.2020)
In the movie, to test AVA’s response ability, Caleb tells idioms and complicated sentences
like “We have to break the ice…” which would be difficult for an ordinary robot to respond.
Howeever AVA responds immediately, with no hesitation. At this point, however, it should be
noted that when Caleb asks if she knows what ‘breaking the ice’ means, Ava responds with a
dictionary definition and says: “Overcoming initial social awkwardness”.
To determine that, Nathan uses an assessment close to Touring Test to Ava applied by
Caleb. Since Caleb knows that Ava is not a human, the object of the test similar to Touring Test
in the movie, is not to tell whether she is a human or a machine, but to determine if she has
DİYALEKTOLOG
ULUSAL SOSYAL BİLİMLER DERGİSİ SİNEM GÜDÜM
YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
98
consciousness as well as ability to understand, empathize and manipulate other people. This is
actually what Nathan really wants to test: Can AVA manipulate Caleb to help her escape?
Here one question comes to mind: Can someone’s humanity be tested on observing the
methods of manipulation one is capable of emposing onto the other? Or does being a human
require something much more than the ability to use some mind manipulation techniques?
When Caleb applies the test to Ava, Caleb comments ‘testing Ava through conversation is
like a closed loop, just like testing a chess program through playing chess’; meaning Ava might
be perfect even better than Caleb in communicating through language (AI can be more intelligent
than humans in the specific programs that are uploaded to them), so to test Ava’s consciousness
through tools of language is not sufficient. This comment of Caleb’s is mentioned again at the
end as Nathan asks Caleb ‘Did we ever get passed that chess problem. Is she feeling or is she
simulating?’ So even though Ava seems to pass the test they applied, the results are considered
unreliable since the testing tools reveal circular references.
2. THE FUTURE OF AI AND CONSUMPTION VIA ANTHROPOMORPHISM
The ‘EX-MACHINA’ movie successfully makes the audience imagine where the future of
AI or ’thinking machine’ is heading. The Cyborg’s thinking patterns are told to be emulated from
the search engines (which show the way people think: chaotic, impulsive, patterned) and uploaded
to Ava’s gel based brain. The movie answers ‘yes’ to several questions about strong AI: Ava is
able to understand and think, have free will, make decisions, learn on her own and modify her
behavior. Example of Ava’s free will in decision making is when she states it is hard to choose
among so many options of where to go first when she is free, though then making her subjective
choice. Ava also states she searches the web for areas of her interest, showing she can learn on
her own.
Another question about AI is ‘will the humanoid robots be able to consciously experience
‘life’ like humans?’ One may start by defining what consciousness is here. Consciousness brings
along sentience( the capacity to feel, perceive or experience subjectively), and awareness (having
a sense of selfhood, and the executive control system of the mind). It must be noted that high
intelligence does not necessarily mean consciousness, but Ava seems to have a high IQ too…
She does smile at the end of the movie, showing her ‘feeling’ of happiness after she finally gets
free and sees the sun! Another example of a conscious act by one of Nathan’s humanoid
prototypes is seen in the character Kyoko, who has been unresponsive and servile for the whole
film, but at the end she stabs Nathan. Supposedly, she wants to escape from her ‘miserable’ life,
which surely is an act of personal awareness (guessing that Kyoko could not be programmed by
Nathan to kill him).
As one may clearly see, the ‘awareness’ escorted by the ‘free will’ of the humanoid is being
introduced to the audience through out the ‘Ex-Machina’ movie. This kind of anthropomorphism
regarding the ‘feeling machines’ is also very profitable for the promotion of the robot agenda in
Using Anthropomorphism in Movies For DİYALEKTOLOG
Promoting Humanoid Robots in The Age of ULUSAL SOSYAL BİLİMLER DERGİSİ
Artificial Intelligence: YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
Ex-Machina Movie
99
today’s world. In the below advertisement, made for the GM hundred thousand mile warranty;
one may clearly see such humanization of the ‘working’ robot, that chooses to end its ‘life’ by his
‘freewill’, just like any other failing human could choose to do (Please read the QR code via a
smart phone to watch the advertisement)
(https://www.youtube.com/watch?v=B3NGN4t4hm4 11.05.2020)
CONCLUSION
As Robert (2017:1) claims, research has shown that humans are more likely to trust and
feel safe with robots they view as more human. However, there are also some drawbacks to such
humanization of robots. Some of the problems with humanizing robots can be stated as follows:
a. Humanization impacts our emotions towards robots. Such emotions are likely to range
from feelings like “fondness” when robots are less humanized to stronger emotions such as “love”
when robots are humanized.,
b. Humanization impacts our expectations. The more humanized robots are, the more one
expects them to take on human responsibilities.
c. People’s expectations on how, and on what level the others will be able to interact with
robots increase.
Riek and friends (2009: 246) stated that while people do not have any reason to hesitate
when abusing property within the solitude of their home, they can not do the same to another
human. That kind of aggressive behavior would definitely be morally unacceptable. The results
of Riek and friends’ experiment showed that people actually empathized nearly as much with any
humanoid robot as they did with another human. Thus, witnessing a crime made against a
humanoid robot, could bring emotional harm to the people watching it. However, the results of
the experiment also showed that people would not feel as bad if the robot was mechanical looking.
So, in short, the way a robot looks is actually very important and it makes a difference when it
comes to designing ethical frameworks for the digital age. Advertisers, together with the brand
ambassadors online know this very well, and thus they use anthropomorphism to its limits for
keeping the ‘subject’ consume more, until he/she starts consuming his/her own identity, and
become presented by a self-made avatar online.
DİYALEKTOLOG
ULUSAL SOSYAL BİLİMLER DERGİSİ SİNEM GÜDÜM
YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
100
REFERENCES
Adorno, T. & Horkheimer, M. (2019). Towards a New Manifesto (Translated by: Rodney
Livingstone). London: Verso Books
Ascott, R.(2016). 35 yrs ago… Connectivity: Art and Interactive Telecommunications (Online:
https://www.facebook.com/notes/roy-ascott/35-yrs-ago-connectivity-art-and-
interactive-telecommunications-roy-ascott/10153596682076073 ) (23.02.2020)
Bilgin, N. (2011). Eşya ve İnsan. İstanbul: Gündoğan Yayınları
Caporael, Linnda R. (1986), “Anthropomorphism and Mechanomorphism: Two Faces of the
Human Machine”, Computers in Human Behavior, 2 (3), 215–34.
Chandler, Jesse and Norbert Schwarz (2010), “Use does not wear ragged the fabric of friendship:
Thinking of objects as alive makes people less willing to replace them”, Journal of
Consumer Psychology, 20 (2), 138–45.
Daws, R. (2015).“Google's recruitment process is straight from Ex Machina (Online)”
http://www.developer-tech.com/news/2015/aug/27/googles-recruitment-process-
straight-ex-machina/ (24.04.2020)
Goetz, Jennifer & Kiesler, Sara & Powers, Aaron. (2003). Matching robot appearance and
behavior to tasks to improve human-robot cooperation. Proceedings - IEEE International
Workshop on Robot and Human Interactive Communication. 55 - 60.
10.1109/ROMAN.2003.1251796.
Guthrie, Stewart (1993), Faces in the Clouds: A New Theory of Religion, New York: Oxford
University Press.
Hinds, P.J., Roberts, T.L., & Jones, H. (2004). Whose job is it anyway? A study of human-robot
interaction in a collaborative task. Hum.-Comp. Interact., 19(1)
Ho, C.C., MacDorman, K.F., & Pramono. Z.A.(2008). Human Emotion and The Uncanny Valley:
A GLM, MDS, and Isomapanalysis Of Robot Video Ratings. ACM/IEEE HRI:
Klein, A.G., Helfgen,J.,& Pfeifer, K.,(2013),"Humanized Products in Tv Ads: How
Anthropomorphism Can Elicit Emotions, Enhance Attitudes, and Affect Purchase
Likelihood",E-European Advances in Consumer Research,Volume 10, eds. Gert
Cornelissen, Elena Reutskaja, and Ana Valenzuela, Duluth, MN : Association for
Consumer Research, 265-266.
Krach, S., Hegel, F., Wrede, B., Sagerer, G., Binkofski, F., & Kircher, T. (2008). Can Machines
Think? Interaction and Perspective Taking with Robots Investigated. fMRI.PLoS ONE,
3(7), 101
Using Anthropomorphism in Movies For DİYALEKTOLOG
Promoting Humanoid Robots in The Age of ULUSAL SOSYAL BİLİMLER DERGİSİ
Artificial Intelligence: YAZ SUMMER 2020-SAYI/ ISSUE 24-SAYFA PAGE 93-101
Ex-Machina Movie
101
Merriam-Webster dictionary (2020):https://www.merriam-webster.com/dictionary/humanization
(02.05.2020)
Mori, M. (2012). The Uncanny Valley (Translated by Karl F. MacDorman and Norri Kageki).
IEEE Robotics & Automation Magazine. 98-102
Rafaeli, S.(1988). Interactivity: From New Media to Commnnication. SAGE Annual Review of
Communication Research, 110-134
Reavey,B. Puzakova, M., Andras, T.L., & Kwak, H. (2018) The Multidimensionality of
Anthropomorphism in Advertising: The Moderating Roles of Cognitive Busyness and
Assertive Language, International Journal of Advertising, 37:3, 440-462
Riek, L. D, Rabinowitch, T.C, Chakrabarti, B & Robinson, P. (2009), “How Anthropomorphism
Affects Empathy Toward Robots,” in HRI ’09: Proceedings of the 4th ACM/IEEE
international conference on human- robot interaction, ed. François Michaud, Matthias
Scheutz, Pamela Hinds, and Brian Scassellati, CA: ACM, 245–46.
Robert, L. P. (2017). The Growing Problem of Humanizing Robots. International Robotics &
Automation Journal, 3(1), Available at SSRN: https://ssrn.com/abstract=3027628
(23.05.2020)